Planta DOI 10.1007/s00425-014-2219-7

ORIGINAL ARTICLE

Molecular identification and functional analysis of a maize (Zea mays) DUR3 homolog that transports urea with high affinity Guo-Wei Liu • Ai-Li Sun • Di-Qin Li • Asmini Athman • Matthew Gilliham • Lai-Hua Liu

Received: 4 June 2014 / Accepted: 3 December 2014  Springer-Verlag Berlin Heidelberg 2014

Abstract Main conclusion Successful molecular cloning and functional characterization of a high-affinity urea permease ZmDUR3 provide convincing evidence of ZmDUR3 roles in root urea acquisition and internal urea-N-remobilization of maize plants. Urea occurs ubiquitously in both soils and plants. Being a major form of nitrogen fertilizer, large applications of urea assist cereals in approaching their genetic yield potential, but due to the low nitrogen-use efficiency of crops, this practice poses a severe threat to the environment through Electronic supplementary material The online version of this article (doi:10.1007/s00425-014-2219-7) contains supplementary material, which is available to authorized users. G.-W. Liu  A.-L. Sun  L.-H. Liu Department of Plant Nutrition and Key Laboratory of Plant-Soil Interactions, Ministry of Education, College of Resources and Environmental Sciences, China Agricultural University, Beijing 100193, China Present Address: G.-W. Liu Institute of Plant Biology, University of Zurich, 8008 Zurich, Switzerland D.-Q. Li (&)  L.-H. Liu (&) Key Laboratory of Crop Physiology and Molecular Biology, National Education Ministry, Hunan Agricultural University, Changsha, Hunan 410128, P. R. China e-mail: [email protected] L.-H. Liu e-mail: [email protected] A. Athman  M. Gilliham ARC Centre for Plant Energy Biology School of Agriculture, Food and Wine Waite Research Institute, University of Adelaide, PMB 1, Glen Osmond, SA 5064, Australia

their hypertrophication. To date, except for paddy rice, little is known about the biological basis for urea movement in dryland crops. Here, we report the molecular and physiological characterization of a maize urea transporter, ZmDUR3. We show using gene prediction, PCR-based cloning and yeast complementation, that a functional fulllength cDNA encoding a 731 amino acids-containing protein with putative 15 transmembrane a-helixes for ZmDUR3 was successfully cloned. Root-influx studies using 15N-urea demonstrated ZmDUR3 catalyzes urea transport with a Km at *9 lM when expressed in the Arabidopsis dur3-mutant. qPCR analysis revealed that ZmDUR3 mRNA in roots was significantly upregulated by nitrogen depletion and repressed by reprovision of nitrogen after nitrogen starvation, indicating that ZmDUR3 is a nitrogen-responsive gene and relevant to plant nitrogen nutrition. Moreover, detection of higher urea levels in senescent leaves and obvious occurrence of ZmDUR3 transcripts in phloem-cells of mature/aged leaves strongly implies a role for ZmDUR3 in urea vascular loading. Significantly, expression of ZmDUR3 complemented atdur3-mutant of Arabidopsis, improving plant growth on low urea and increasing urea acquisition. As it also targets to the plasma membrane, our data suggest that ZmDUR3 functions as an active urea permease playing physiological roles in effective urea uptake and nitrogen remobilization in maize. Keywords Molecular function  Maize  Nitrogen uptake and remobilization  Urea transporter ZmDUR3 Abbreviations NUE Nitrogen-use efficiency N Nitrogen AN Ammonium nitrate

123

Planta

RO water FM4-64

SD PM

Reverse osmosis water N-(3-Triethylammoniumpropyl)-4-(pdiethylaminophenyl-hexatrienyl) pyridinium dibromide Synthetic dextrose Plasma membrane

Introduction Maize (Zea mays) is one of the Earth’s most widely cultivated crops. However, inefficient acquisition and use of soil nitrogen (N) by cereal crops, including maize, result in a fairly low N-use efficiency (NUE); reported values are only 30–50 % of the applied N-fertilizer (Tilman et al. 2002; Andrews and Lea 2013; Wang et al. 2013). This has severe problems including higher production cost, ecosystem impairment, water and air pollution (Cameron et al. 2013). Thus, as a consequence, manipulating the biological processes of root N-acquisition/use to better the crop NUE has long been regarded as a crucial aspiration that will lead to minimizing the expense of crop fertilization and its impact on environmental quality. Urea is applied in a large amount as a N-fertilizer in modern crop production because its N is readily accessible to plants, it has a high N content and low manufacturing cost, as such it amounts to [50 % of the world N-fertilizer consumption (Kojima et al. 2007; Wang et al. 2008; Witte 2011). With respect to the accessibility of urea-N to the plant, a general suggestion is that most of urea supplied to crops is predominantly taken up as NH4? due to urea’s fast hydrolysis by microbial-derived urease activity in soils, and that direct urea capture and internal degradation by plant cells may not be prominent (Marschner 1995). However, in the recent years, physiological and molecular aspects of root uptake and utilization of organic N such as urea, amino acids and small peptides have received greater attention (Kojima et al. 2007; Rentsch et al. 2007; Merigout et al. 2008; Wang et al. 2012). Since organic N represents a major part (approximately 90 %) of the total soilN (Yu et al. 2002; Jones et al. 2005), their effective capture and use by the plants is believed to contribute environmentfriendly organic farming. The physiological potential for direct absorption by higher plants of urea as N source has been documented for many years (Krogmeier et al. 1989), but how urea moves into root cells has only been revealed recently. Short-time influx studies using 15N-urea showed the existence of a high- and low-affinity transport system for urea in roots of Arabidopsis and rice plants (Kojima et al. 2007; Wang

123

et al. 2012). Accordingly, molecular cloning and the functional characterization of high-affinity transporter DUR3 homologs from Arabidopsis and rice has revealed that urea can be imported through a protein-mediated and energy-dependent process when external urea is in the low micromolar range, which are concentration comparable to those detected in most agricultural soils (Liu et al. 2003a; Wang et al. 2012). The affinity of AtDUR3 and OsDUR3 for their substrate was, respectively, measured at approximately 3 or 10 lM when expressed in Xenopus oocytes (Liu et al. 2003a; Wang et al. 2012), which is comparable to the values reported for urea uptake by the roots of Arabidopsis and rice (Kojima et al. 2007; Wang et al. 2012). Transcriptional upregulation of both AtDUR3 and OsDUR3 in the roots was triggered by N deficiency and presence of their substrate (Kojima et al. 2007; Wang et al. 2012), indicative of their physiological significance in relation to plant N nutrition. When overexpressed in Arabidopsis, OsDUR3-harbouring lines displayed a marked growth improvement on (sub-)millimolar urea and a great enhancement of root urea uptake, whereas a strong knockdown of OsDUR3 transcripts by T-DNA insertion caused obvious growth inhibition and less urea absorption of rice on urea (Wang et al. 2012). The ability of higher plants to recycle/(re-)use nutrients, e.g., N sources, may confer an economical benefit. Several studies suggest that urea generation from arginine by arginase action in the ornithine cycle would allow plants to reutilize N from protein breakdown, particularly during seed germination or tissue senescence (Wang et al. 2008; Witte 2011). Indeed, urea content could be determined at a significantly higher level in aged rice leaves compared to that in young ones despite the occurrence of a relatively stable endogenous urease activity (Wang et al. 2012). In such a case, mobilization of urea as an N source to growing tissues might be required, and this seems to occur. Experimental evidence showed that the rice urea transporter OsDUR3 mRNA increased markedly from day 1–6 after seed germination as well as being positively correlated to a higher urea level in the old leaves (Wang et al. 2012). This suggests an additional physiological role of OsDUR3 in plant N-remobilization and/or -reuse. However, whether or not rice DUR3 is involved in the longdistance vascular transport of urea needs to be further confirmed. To date, the molecular bases of urea transport systems in mammals and microbes have been reported for a long time, but knowledge about urea movement and assimilation in higher plants is still limited. In the present work, we report the molecular identification and physiological characterization of a high-affinity urea permease ZmDUR3 in maize (Zea mays). Using approaches including gene prediction in

Planta

silico, PCR-based cloning and heterologous growth complementation test, a putative full-length cDNA of a maize DUR3 homolog that is functional for urea transport in yeast was cloned. Moreover, gene expression studies using quantitative as well as in situ RT-PCR, GFP-based protein subcellular localization, transgenic approach and urea influx assays were performed to decipher the crucial physiological function of ZmDUR3 in urea uptake and remobilization in maize plants.

Materials and methods Plant culture Seeds of maize inbred line B73 (Z. mays L. National Maize Improvement Centre, http://cab.cau.edu.cn/caumaize/tea cherlist.php) were sterilized with 10 % (v/v) H2O2 for 20 min, rinsed with distilled water 5 times, soaked in saturated CaSO4 for 6 h, and thereafter germinated in the dark on humid filter paper at room temperature for 2–3 days. Sprouting seeds were transferred to quartz–sand immersed in water until seedlings showed two leaves. Young plants with a similar size had there endosperm removed and were used for hydroponic experiments. Porcelain pots contained 2.2 L solution which was used to grow three plants each (in a growth chamber, 14 h light, 28 C day, 22 C night). The photosynthetic photon flux density was 250–300 mmol m-2 s-1 at the top of the plant canopy during the 14 h light period. The basic medium contained 0.25 mM KH2PO4, 0.75 mM K2SO4, 0.65 mM MgSO4, 0.1 mM KCl, 200 lM Fe-EDTA, 1 lM H3BO4, 0.10 lM CuSO4, 1 lM ZnSO4, 0.10 lM (NH4)6Mo7O24, 1 lM MnSO4, pH 6.0 (adjusted using Tris-buffer); 2 mM Ca(NO3)2 was supplied as the N source. The nutrient solution (using Milli pore water, Merck) was refreshed every 2 days and aerated by an electric pump. For N-dependent gene expression analysis, maize plants were hydroponically cultivated in the basic solution for 9, 10, or 11 days. Except for the control treatment, that had a constant supply of the basic solution containing 2 mM Ca(NO3)2, the plants were transferred to N-free solution for 1, 2, or 3 days (replacing the Ca(NO3)2 with 2 mM CaCl2). After 3 days of N-starvation (note: the solution was refreshed everyday), plants were resupplied with 2 mM N in the form of urea or (NH4)2SO4 or KNO3 or glutamine for 3 h. No ammonium (see the assay method below) in ureainduction solution was detectable during a period of urea resupply, indicating that urea may not be hydrolyzed in the treatment solution. Roots and shoots were harvested separately, frozen in liquid nitrogen immediately and stored at -80 C for total RNA isolation and gene expression analysis.

To check whether or not urea would be degraded during its 3-h supply in the gene expression study, ammonium was measured using ion-selective electrodes (WP-90) method provided by the manufactory protocol (TPS Pty Ltd, Brisbane, Australia). 30 mL nutrient solution was sampled at a time point of 1 or 2 or 3 h after plants were subjected to 2 mM urea induction, and measured by WP-90 apparatus. A set of known ammonium concentrations (0–250 lM, using NH4Cl) was prepared in the basic nutrient solution (without N, see above ‘‘basic medium’’) and used to generate a standard curve for the calculation of ammonium concentrations of the samples. At least three samples at every sampling time were tested. Ammonium in the ureacontaining solution without maize plant was also measured in the same manner. In all cases, ammonium was not measureable in the urea-containing samples. For analysis of urea content as well as ZmDUR3 expression in maize leaves at different ages, plants were cultivated for 11 weeks in pots (0.20 m diameter, 0.35 m height) filled with 5 kg agricultural soil, and the above nutrient solution (1 L) was applied at a elongation and tasselling stage. Leaves were washed three times with distilled water and harvested in liquid nitrogen immediately for later experiment use. Urea content was measured as described in Wang et al. (2012). Gene isolation and yeast functional complementation A putative ZmDUR3 ORF was amplified by Pfu polymerase (Invitrogen) using PCR. Gene-specific primers containing BamHI and HindIII site, respectively, were used (50 -TTAggatccATGGCCGCTGGCGGCGCC-3, 5-CGCaa gcttTTAAGCTAGCGAAAG ATTATC-30 ). The resulting PCR product was cloned into pGEM-T Easy vector (Promega), sequenced and checked for its correction compared with the maize genomic database. The ZmDUR3 ORF was sub-cloned into a yeast expression vector pHXT426 (Wieczorke et al. 1999) using BamHI and HindIII sites. Yeast transformation and complementation were performed as described in Liu et al. (2003a). Transformants were first grown on SD agar (Oxid) medium (uracil-deficient yeast nitrogen-base. Difco) containing 20 mM NH4? [i.e., 10 mM (NH4)2SO4] as the N source. A single colony from each transformation was picked for the growth complementation assay on urea as the only N source. The medium pH was adjusted by 1 M HCl or KOH. RNA extraction and gene expression analysis by qPCR Quantitative RT-PCR (qPCR) was applied to determine the expression of ZmDUR3 in maize roots. Total RNA was isolated using TRIzol reagent (Sangon). DNaseI (Invitrogen) was used for removal of DNA contamination in RNA

123

Planta

samples. First strand cDNA synthesis was carried out on 2 lg total RNA as a template using M-MLV Reverse Transcriptase (Promega) according to the manufacturer’s protocol. qPCR was conducted in 20 lL volume (containing 2 lL of 1:10 diluted original cDNAs, 200 nM of each gene-specific primer, and iQTM SYBR Green Supermix from Bio-Rad) using a Bio-Rad iCycler. PCR cycling parameters were set as following: initial denaturation at 95 C for 30 s, then 42 cycles of 95 C for 5 s, 60 C for 15 s, 72 C for 10 s, and a final extension at 72 C for 7 min. The following primers were used for qPCR: ZmDUR3-F, 50 -CTTCAAGGGCTCCTACCTCA-30 and ZmDUR3-R, 50 -CATACCAGCC CACCCAGAAG-30 . ZmTubulin (ZmTUB4, as an internal control) forward primer 50 -GCTAT CCTGTGATCTGCCCTGA-30 and reverse primer 50 -CGCCAAACTTAATAACCCA GTA-30 (Gu et al. 2012). ZmGAPDH (glyceraldehyde-3-phosphate dehydrogenase gene, as a second reference) forward primer 50 -CTGGTTTCTACCGAGTTCCTTG-30 and reverse primer 50 -CGGCATACACAAGCAGCAAC-30 (Gu et al. 2012). Localization of GFP-tagged ZmDUR3 in Arabidopsis roots ZmDUR3 ORF without its stop codon was amplified by Pfu (Invitrogen) using the following primers containing a BglII site (50 -CCggatccATGGCCGCTGGCGGC GCCGGC-30 and 50 -ATggatccAAGCTAGCGAA AGATTATCT-30 ), and cloned into a vector pCF203 [carrying a CaMV 35Spromoter (35Spro), GFP gene, rbcs-terminator (rbcsterm), spectinomycin-resistant marker and kanamycin-resistant marker gene; Liu et al. 2003a], yielding a construct ‘‘pCF203-35Spro:ZmDUR3-GFP:rbcsterm’’. Arabidopsis Col-0, atdur3-1 or atdur3-3 plants (Kojima et al. 2007) were transformed by dipping inflorescences into a cell suspension (OD600 = 0.6) of Agrobacterium GV3101 harbouring the above construct. Transgenic plants (selected by 50 lg L-1 kanamycin) were grown for 8 days on vertically placed agar medium containing half-strength Murashige and Skoog (MS) medium and were used for a microscopic assay. As a plasma membrane protein marker, OsDUR3-GFP transformed Arabidopsis line was used as a reference (Wang et al. 2012). For staining of the plasma membrane, plant roots were incubated with 20 lM FM4-64 (Molecular Probes, N-(3-triethylammoniumpropyl)-4-(p-diethylaminophenylhexatrienyl) pyridinium dibromide) for 10 min, washed three times with sterile water before observation with the confocal microscope. The roots’ cells were scanned with an energy excitation at 488 or 543 nm by confocal laser scanning microscope (Eclipse TE2000-E, Nikon, Tokyo, Japan).

123

Expression analysis by in situ PCR on maize leaves For preparing plant material, seeds of maize inbred line B73 were bubbled in RO water for 6 h and placed under humid pebbles for 4 days to germinate. Once roots were 3 cm long, they were grown in porcelain pots containing 2.3 L of basic medium with nutrient compositions as described in ‘‘Plant culture’’. The nutrient solution was changed every 2 days and continuously aerated by an electric pump. Leaves from 14-day grown plants were harvested for fixation and later for in situ PCR analysis. In situ PCR was conducted according to the method described in Munns et al. (2012) but with the following modifications. 70 lm sections were obtained using a VT 1200S Vibrating Microtome (Leica) and a paintbrush was used to transfer sections to 0.2 mL tubes containing 100 U of RNAseOUT (New England Biolabs) in 100 lL cold sterile water. DNase in the samples was inactivated by adding EDTA to a final concentration of 15 mM and then heating at 70 C for 15 min. The reverse primer (see below) was used as a gene-specific primer for cDNA synthesis using Thermoscript RT (Life Technologies). PCR cycling parameters were: initial denaturation at 98 C for 30 s, then 30 cycles of 98 C for 10 s, 57 C for 30 s, 72 C for 12 s and a final extension at 72 C for 10 min using Phusion Polymerase (New England Biolabs). The sections were developed for 15 min. The sequences of the gene-specific primers used for in situ PCR are: Hv18S, forward 50 -GGTAATTCCAGCTCCAAT -30 , reverse 50 -G TTTATGGTTGAGACTAG-30 ; ZmDUR3, forward 50 - GT CGACAACGGCTACTGGA-30 , reverse 50 -TGTCGTAGG TGCAAAGGGAG-30 . Creation of ZmDUR3-transformed Arabidopsis lines, growth test and 15N urea root uptake ZmDUR3 ORF was synthesized by Pfu (Invitrogen) using two primers containing BglII and SalI sites (50 -AAGagatctATGGCCGCTGGCGGCGCC-30 , 50 -TGGgtcgacTTAA GCTAGCGAAAG ATTATC-30 ), and cloned into a GFPremoved pCF203 (pCF203-) using BglII/SalI ends, yielding a construct, i.e., pCF203-35Spro:ZmDUR3:rbcsterm. Arabidopsis Col-0 and atdur3-3 plants (Kojima et al. 2007) were transformed by the method described above. Harvested seeds were selected on kanamycin (50 lg mL-1)containing 1/2 MS agar medium to obtain transformants. Several independent homozygous ZmDUR3-harboring lines were created in T2 or T3 generation for experimental use. For the growth complementation test, surface-sterilized seeds were grown for 16 days on the 1/50 B5 agar medium without N or containing different concentrations of urea or NH4NO3 with a supplementation of 1 lM NiSO4. To

Planta

confirm the expression of ZmDUR3 in the Arabidopsis transgenic plants, semi-quantitative reverse transcription polymerase chain reaction (semi-qPCR) was performed with total RNA extracted from Col-0, atdur3-3 and transgenic lines (Liu et al. 2003a). Primers used for the semiqPCR were: ZmDUR3, 50 -GTCGACAACGGCTACTG GA-30 and 50 - TCTCCACCGTGGTGATCTG-30 ; AtACT2, 50 -TCCAAGCTGTTCTCTCCTTG-30 and 50 -AGACGGA GGATGGCATGAG-30 (Liu et al. 2003a). Roots urea uptake study was performed with atdur3-3 and its ZmDUR3-overexpessing lines cultivated hydroponically for 3 weeks. Tracer urea with 98.65 % 15N abundance (Novachem PTY Ltd., Melbourne, Australia) was applied, and samples of ca 3–4 mg (dry plant material) were used to determine 15N-content using mass spectrometry (Sercon, Hydra 2020, Crewe, UK). A set of urea concentrations (i.e., 5, 10, 20, 30, 50 and 200 lM) was designed for the uptake kinetic study. For the detailed protocol of 15N urea root uptake refer to the description in Wang et al. (2012).

cDNA or ORF was predicted that contained 2,124 bp and coding for a 707-amino acid peptide with 76.4 % identity to OsDUR3. Based on this prediction, we designed primers for the amplification of the 2,124 bp cDNA. Consequently, a reverse-transcriptional polymerase chain reaction (RTPCR) and DNA sequencing allowed us to isolate a cDNA with 2,196 bp rather than 2,124 bp as predicted. Thus, we termed this maize 2,196 bp cDNA as the full-length ZmDUR3 (Accession Number KM271989). In comparison with its corresponding genome sequence in maize (B73), the ZmDUR3 coding sequence is derived from four exons with a size of 194, 106, 664 and 1,232 bp, respectively (Fig. 1a); its coding protein has features consistent with an integral membrane protein. In silico analysis using ‘‘DNASTARLasergene8’’ revealed that ZmDUR3 contains 731 deduced amino acid residues with a molecular mass of 77.4 kD, shares 82.2 % sequence identity with OsDUR3, and is predicted to possess 15 transmembrane domains with an extracellular N-terminus and a cytoplasmic C-terminus (Fig. 1b, Supplementary Fig. S1).

Results

Heterologous expression of full-length ZmDUR3 restored yeast mutant growth on urea

Identification and molecular cloning of a functional open reading frame encoding the DUR3 homolog from maize To explore the molecular bases of urea movement in dryland cereal crops, we took maize (Z. mays, B73) as a model. The sequence of the active urea transporter from rice, OsDUR3 (Wang et al. 2012), was used as a reference to search for homologs in maize databases (using a BLAST E-value cutoff of 1e-5). Only one locus ZmGSStuc11-1204.11168.1 (B73) consisting of 3919 bp on the chromosome 6 (http://www.plantgdb.org/search/display/data.php? Seq_ID=ZmGSStuc11-12-04.11168.1) homologous to OsDUR3 was obtained, which was predicted [via ‘‘GENSCAN’’ service (http://genes.mit.edu/GENSCAN.html’’)] to contain an open reading frame (ORF) with only 1782 bp. Since a coding region of DUR3 homologous gene from Arabidopsis or rice (AtDUR3, OsDUR3) was reported to be 2,085 or 2,166 bp, respectively (Liu et al. 2003a; Wang et al. 2012), the 1,782 bp cDNA predicted from the ZmGSStuc11-12-04.11168.1 for ZmDUR3 might not be a complete full-length ORF (i.e., lacking of a N-terminal part). The genomic sequence of maize is now available (Schnable et al. 2009; http://www.maizegdb.org), and this allowed us to extract a 10 kb genomic DNA (i.e., from Chr.6: 104450983…120850613; http://www.plantgdb.org/ ZmGDB/cgi-bin/getRecord.pl?dbid=1&resid=9&chrUID= 16352) containing the ZmGSStuc11.-12-04.11168.1 and by analyzing this using the GENSCAN service, a putative

The functionality of full-length ZmDUR3 for urea transport was tested using a growth complementation test of a urea transport yeast mutant YNVW1 (Liu et al. 2003a). Transformation of YNVW1 with the yeast expression vector pHXT426 containing ZmDUR3 driven by a hexosetransporter promoter (Wieczorke et al. 1999) restored the growth of the mutant on 2 mM urea as a sole N source, although the growth rate of the ZmDUR3-harbouring mutant appeared to be slower than that of the wild-type strain 23346C (Mat a, Dura3; a positive control) transformed with empty vector pHXT426 (Fig. 2). No obvious growth was observed for the negative control, i.e., YNVW1 consisting of the empty vector (Fig. 2). Since the urea transport activity of DUR3 from Arabidopsis and rice was reported to occur preferentially at a lower pH (Liu et al. 2003a; Wang et al. 2012), the effect of medium pH on ZmDUR3-mediated urea transport in yeast was assayed. However, no marked growth rate difference of ZmDUR3transformants on 2 mM urea was observed at pH 5 or 7, suggesting that ZmDUR3-facilitated urea import across the yeast plasma membrane (PM) occurred under acidic conditions but with less pH sensitivity. Since DUR3 homologs belong to the sodium-solute symporter super family, we tested if sodium would influence ZmDUR3-assistant urea transport in yeast. Addition of sodium (1–10 mM) in the urea-containing medium was not found to affect the growth of the ZmDUR3-harbouring YNVW1 mutant (data not shown).

123

Planta

a

ZmDUR3

3198 bp

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

exon

b

84 bp

4.0

4.5

83 bp

5.0

5.5

6.0

6.5 bpp

intron

TMHMM posterior probabilities for sequence 1.2

Probability

1 0.8 0.6 0.4 0.2 0

0

200

100

300

Transmembrane Fig. 1 The genomic organization of ZmDUR3 and the predicted membrane topology of the ZmDUR3 protein. a Exon–intron structure of ZmDUR3. The open reading frame of ZmDUR3 cDNA is derived from four exons (indicated in the blue color, with a size of 194, 106, 664 and 1,232 bp, respectively) was cloned and confirmed by

1:1

1:20

1:200

pH

23346C

YNVW1

5

YNVW1+ZmDUR3 23346C

YNVW1

7

YNVW1+ZmDUR3 Fig. 2 Growth complementation of a yeast mutant by heterologous expression of ZmDUR3. A yeast mutant YNVW1 (dur3, ura3) disrupted in the DUR3 gene based on the strain 23346C (Mat_a, ura3) (Liu et al. 2003a) was not able to grow on 2 mM urea as a sole N source (Liu et al. 2003a). The mutant was transformed with a yeast expression vector pHXT426 alone (as a negative control) or harbouring the ZmDUR3 ORF, and the 23346c strain transformed with the empty vector serves as a positive control. Transformants were grown first on SD medium (see Materials and methods) containing 20 mM NH4? as the N source for 4 days, then cells from a single colony were picked, suspended (in 100 lL water), diluted (in a series of 1:1, 1:20, 1:200) and spotted (1 lL) onto the SD medium consisting of 2 mM urea as the only N source at pH 5 or 7. Pictures were photographed after the growth of yeast for 5 days in the urea medium

123

400

inside

500

600

700

outside

sequencing. b Hydropathy analysis of the ZmDUR3 peptide using ‘‘CBS prediction servers’’ (http://www.cbs.dtu.dk/services/). ZmDUR3 exhibits 15 transmembrane alpha-helices (in red) with the N- and C-terminus outside and inside the cytoplasm, respectively

Subcellular localization of the ZmDUR3 protein To determine protein subcellular targeting, a cytosolic green fluorescence protein gene (GFP) was fused to the C-terminus of ZmDUR3 (35Spro:ZmDUR3-GFP:rbcsterm) and its expression of this protein fusion was driven under the control of the cauliflower mosaic virus (CaMV) 35S promoter; this construct was stably introduced into Arabidopsis (Col-0) plants. The green fluorescence signal in the root cells of several independent transgenic lines (grown for 8 days on the agar nutrient plate) was visualized using confocal microscopy. The green signal derived from the ZmDUR3-GFP overlapped well with the red fluorescence of the plasma membrane (PM)—a signal derived from 10 min incubation with FM4-64 (Tanaka et al. 2008) (Fig. 3). This overlap can clearly be seen in the yellow of the superposition image, which is a consequence of the merged green and red signals (Fig. 3). Furthermore, the localization pattern of ZmDUR3 greatly resembles that of GFP-linked rice OsDUR3 (Fig. 3), which was previously reported to reside mainly at the PM (Wang et al. 2012). Moreover, the smaller size of certain vacuoles compared with that of their corresponding cells could be clearly distinguished (Fig. 3), indicating that the ZmDUR3-GFPderived green signals or the merged yellow signals should not be from the tonoplast but the plasma membrane. As such, the subcellular localization of ZmDUR3 is most likely to be on the PM of plant cells.

Planta

GFP

PM

merged

transmission

PM

V

TP

V

ZmDUR3

PM

FM4-64

V

V

TP

OsDUR3

V

Fig. 3 ZmDUR3 protein is mainly localized to the plasma membrane in transgenic Arabidopsis plants. The 35Spro:ZmDUR3-GFP:rbcsterm construct was introduced into Arabidopsis (Col-0) plants, and cells were visualized via confocal microscopy (see Materials and methods). The cell images presented are from the root elongation zone of 8-dayold transgenic lines. A plasma membrane-localized marker protein OsDUR3 (Wang et al. 2012), which fused with GFP and transformed

in Arabidopsis, serves as a reference. Plasma membranes were stained red with FM4-64 for 10 min (Tanaka et al. 2008). Colors of green, red and yellow derived from GFP, FM4-64 and superposition of green and red, respectively. FM4-64, N-(3-triethylammoniumpropyl)-4-(pdiethylaminophenyl-hexatrienyl) pyridinium dibromide; PM plasma membrane, V vacuole, TP tonoplast

Transcriptional regulation of ZmDUR3 by plant nitrogen status

(Note: ammonium was not detectable in the urea solution during a period of 3-h urea resupply).

Nitrogen-related transcriptional regulation of ZmDUR3 was analyzed by quantitative reverse transcription polymerase chain reaction (qRT-PCR), performed with total RNA from roots of 12-day-old maize plants (B73) grown hydroponically. Multiple biological and technical replicates were conducted to confirm the gene expression pattern. As seen in Fig. 4, levels of ZmDUR3 mRNA remained relatively low under normal growth conditions with 2 mM NO3- supply, but increased progressively and significantly during N-starvation in a period of 1–3 days (Fig. 4), showing that ZmDUR3 was inducible on an expression level following N-deficiency. Furthermore, a 3-h N reprovision to the roots of 3-day N-starved plants with urea, NH4?, NO3- or glutamine (Gln) caused a significant reduction of ZmDUR3 transcripts (to a level similar to that of normal N-grown plants) (Fig. 4). Such similar expression patterns were also observed when another house keeping gene ZmGAPDH was applied as a reference (Supplementary Fig. S2). This points to ZmDUR3 being N-responsive at least at the mRNA level, with transcriptional repression occurring in the presence of various forms of external nitrogen including its putative substrate urea

ZmDUR3 transcripts occurred in specific cell types of leaves To understand a potential physiological role of ZmDUR3 at the cellular level, we investigated its expression pattern in a cell-specific manner using in situ RT-PCR; this was performed on maize leaves of different ages (see Materials and methods). Strong signals from vital cells, indicated by a blue stain, were detected from amplicons of reverse transcribed 18s rRNA, a ‘housekeeping’ gene as a positive control to show that the technique detects transcripts in the cell types in which they are expressed (Fig. 5). Furthermore, no blue signal was detected when ZmDUR3 primers were used for the PCR but a prior reverse transcription step had not been performed as a negative control, indicating that these primers do not pick up genomic DNA in this assay. In the young leaf (i.e., when it has not fully opened), the cDNA of ZmDUR3 was detected mainly in the vascular bundle sheath cells and surrounding the phloem (Fig. 5). In the mature leaf (7 days after being fully open), the blue color indicative of the occurrence of ZmDUR3 transcripts could be obviously detected surrounding all vascular-tissue

123

Planta

ZmDUR3 relative expression (ZmDUR3/ZmTUB4)

4.5

d

4.0 3.5

c

3.0 2.5 1.5

b

b

2.0

b

ab

ab

a

1.0 0.5 0.0 1d

NO3-

2d -N

Fig. 4 Transcriptional regulation of ZmDUR3 by nitrogen in maize roots. Relative mRNA accumulation of ZmDUR3 was measured by qPCR, which was performed with total RNA from 12-day-old maize roots (B73). Maize cultivation and N treatments are described in the ‘‘Materials and methods’’. Gene-specific primers designed from the partial sequence of the second and third exon of ZmDUR3 were used. Means of three biological replicates and two technical repeats ± SE (n = 6) are shown. Different letters over the bars indicate statistically

3d

Urea

NO3-

Gln

-N 3d

significant differences (P \ 0.05, by one-way ANOVA). Maize tubulin4 gene (Gu et al. 2012) served as an internal reference. The mRNA level of ZmDUR3 in maize root during normal N-treatment (i.e., NO3–) was set to 1 for the calculation of relative gene expression. –N, N-starvation; -N3d, N-starvation for 3 days. N-resupply with urea, NH4?, NO3- or Gln for 3 h. Another constitutive expressing gene, i.e., ZmGAPDH (Gu et al. 2012) was also used as a reference to confirm the gene expression pattern (see Suppl. Fig. S2)

Young leaf Midrib

NH4+

Mature leaf Vein

Midrib

Vein

18S 50 m

50 m

50 m

50 m

ZmDUR3

50 m

50 m

50 m

50 m

ZmDUR3 (no RT) 50 m

50 m

X V

BS

P

BS P

50 m

X

Fig. 5 Cell-specific localization of ZmDUR3 mRNA by in situ PCR on leaves of maize (B73). Plants were grown in hydroponics for 14 days. Leaves were harvested at different developmental stages, i.e., ‘‘young leaf (not fully opened)’’ and ‘‘mature or old leaf (7 days after fully opened)’’. Cells in which transcripts were present stain blue, and ZmDUR3 was detected in various vascular cells, depending

123

50 m

on the treatment and stage. 18S rRNA was used as a positive control to show the presence of cDNA in all vital cells and a no reverse transcription (i.e., no cDNA as negative control) was included to demonstrate lack of genomic DNA contamination. BS vascular bundle sheath, P phloem, V vessel, X xylem

Planta

cells (i.e., those surrounding the phloem, xylem and the bundle sheath cells) (Fig. 5). Accumulation of urea and ZmDUR3 expression in leaves showed age dependency In some circumstances, urea can be generated in the plant via the degradation of N-containing micro-molecules such as proteins and nucleotides during seed germination and senescence, especially in older leaves (Witte 2011). Thus, the expression of ZmDUR3 and its urea contents were quantified in maize leaves through time. The amount of urea in different leaves could be measured in a range from 0.13 to 0.22 lmol per gram fresh material (Fig. 6), but the level of urea in old leaves (e.g., the eighth leaf counted down from the top leaf) was [50 % higher than that of the young leaves (e.g., the second as well as the fifth leaves counted down from the top), where urea remained at similarly low levels (0.13–0.14 lmol g-1 FW) (Fig. 6). Similarly, the level of ZmDUR3 transcript, measured using

b

Urea content ( mol g-1 FW)

a

ZmDUR3 relative expression (ZmDUR3/ZmTUB4)

b

a a

YL

ML

OL

b a

YL

ab

ML

OL

Fig. 6 Measurement of urea content and relative gene expression of ZmDUR3 in differently aged leaves of maize. Plants (B73) were grown in soil for 11 weeks under normal N supply (see Materials and methods). a Urea accumulation in leaves. Detection of urea is as described in Wang et al. (2012). b Relative expression level of ZmDUR3 in the leaves. qRT-PCR was performed as described in ‘‘Materials and methods’’. Gene-specific primers for ZmDUR3 and a housekeeping gene Tubulin were used as stated in Fig. 4. Means of four biological replicates ± SE (n = 4) are plotted. Different letters above the bars showed statistically significant differences (P \ 0.05, by one-way ANOVA). YL young leaf (the second leaf counted down from the top leaf), ML mature leaf (the fifth leaf counted down from the top, which wrapped a cob), OL old leaf (the eighth leaf counted down from the top)

qRT-PCR, displayed a tendency for higher gene expression in the old leaves (Fig. 6). ZmDUR3 overexpression enables better growth of Arabidopsis on low urea and mediates a high-affinity uptake of urea into Arabidopsis roots It has previously been shown that both Arabidopsis wild type and atdur3 plants could not use well low concentrations of urea (e.g., \1 mM) as a sole N source for their growth as compared with the plants on NH4NO3 (Kojima et al. 2007; Wang et al. 2012). To examine the molecular and physiological significance of ZmDUR3 for urea acquisition and use in plants, an expression construct of ZmDUR3 driven by the CaMV 35S promoter was made and subsequently introduced into Arabidopsis wild-type (WT, Col-0) or atdur3-3 plants (Kojima et al. 2007); this yielded several independent ZmDUR3-harbouring homozygous lines for physiological analyses. Semi-quantitative RTPCR was applied to confirm the expression of ZmDUR3 in the different transgenic lines (Fig. 7a), and two representative lines of ZmDUR3-transformed WT and atdur3-3 were used (Fig. 7a). Plants were grown for 16 days on sterile agar nutrient medium supplied with no N, NH4NO3 or urea at different concentrations. Both WT and atdur3-3 plants on low urea (0.5 mM) or no N exhibited a stunted growth phenotype characteristic of significantly reduced biomass as well as N-deficiency symptoms as indicated by anthocyanin accumulation leading to brown/red–brown leaves (Supplementary Fig. S3, Fig. 7b). However, all ZmDUR3 transformants showed normal or improved growth with less N-deficiency symptoms on low urea (Fig. 7b). Accordingly, measurement of a 20-min uptake of 15 N-labelled urea at 50 or 500 lM into roots of 3-week-old plants showed more than 50 % higher urea absorption in the ZmDUR3-transgenic lines than that in the atdur3 mutant (Fig. 7c). At higher urea, e.g., 3 mM or NH4NO3 (0.5 mM) provision, no apparent phenotypic growth deference between WT, mutant and ZmDUR3-expressing lines was observed (Fig. 7b). Moreover, to characterize ZmDUR3 facilitated urea transport kinetics, the experiment of a short-term 15N-urea influx into roots of 22-day-old atdur3-3 mutant plants and a ZmDUR3-transformed line (atdur3-3?ZmDUR3-2) was conducted after starving them of nitrogen for 3 days. Upon a 6-min exposure of the roots to the urea-containing nutrient solution, concentration-dependent urea influx suggested that ZmDUR3-mediated external urea import into the roots displayed enzymatic Michaelis–Menten kinetics (Fig. 8). The root uptake by ZmDUR3 was saturated at *100 lM urea, and exhibited a maximal transport activity, i.e., Vmax at 1.286 ± 0.181 lmol urea g-1 DW h-1 as well as a Km of 9.429 ± 0.816 lM for urea.

123

Planta

a WT

WT +ZmDUR3-1

WT +ZmDUR3-2

atdur3-3

atdur3-3 atdur3-3 +ZmDUR3-2 +ZmDUR3-3

ZmDUR3 AtACT2

b no N

0.5 mM urea

0.5 mM AN

c

Urea uptake ( mol g-1 DW h-1)

3 mM urea

atdur3-3 atdur3-3 +ZmDUR3-2 atdur3-3 +ZmDUR3-3

50

500 Urea concentration ( M)

Fig. 7 Overexpression of ZmDUR3 in Arabidopsis improved plants growth and urea uptake at low concentrations of urea supplied as a sole N source. The ZmDUR3 ORF cloned after the CaMV 35Spromoter (yielding the construct 35Spro:ZmDUR3:rbcsterm) was introduced into Arabidopsis WT (Col-0) or DUR3-defected mutant atdur3-3 (Kojima et al. 2007). Two of several independent transgenic lines of WT ? ZmDUR3 or atdur3-3?ZmDUR3 were shown. a Gene expression test of ZmDUR3 in different lines. Semi-quantitative RTPCR (Liu et al. 2003a) was performed with total RNA from the whole plants cultivated on agar nutrient media as shown in b. Primers for the amplification of ZmDUR3 ORF were applied. Arabidopsis ACTIN2 (AtACT2) served as a reference. b Growth phenotyping of ZmDUR3-

overexpressing lines, WT and atdur3-3 on agar medium containing urea or NH4NO3 (AN) as N sources. c Increased urea uptake was observed in ZmDUR3-expressing Arabidopsis lines. Three-week-old plants of atdur3-3 mutant and its ZmDUR3-transformed lines were used for the 15N-urea uptake study. Roots were exposed for 20 min in nutrient solution consisting of 50 or 500 lM tracer urea (with 98.65 % 15N abundance). 15N incorporated in plants was determined by mass spectrometry and was converted into urea taken up by the roots from medium (see Materials and methods). Data represent mean ± SE (n = 4–5 biological replicates, eight plants in each), and different letters above the bars indicate statistically significant differences (P \ 0.05 by one-way ANOVA). DW plant dry weight

Discussion

detailed understanding of the genetic components responsible for urea transport and their physiological significance related to the effective use of urea have only been demonstrated hitherto in Arabidopsis and rice (Liu et al. 2003a, b; Kojima et al. 2007; Wang et al. 2012, 2013). Here, we report urea transport process of a dryland cereal crop species, i.e., maize at a molecular and physiological level.

Recent physiological and molecular approaches have shown that higher plants like other microorganisms (e.g., fungal and bacterium) are able to directly absorb and utilize urea as a N source to sustain their growth (Kojima et al. 2007; Witte 2011; Wang et al. 2012). However, a

123

Planta

Fig. 8 Concentration-dependent influx of 15N-labeled urea into roots of ZmDUR3-expressing Arabidopsis. Plants grown with 1 mM NH4NO3 for 22 days before transfer to N deficiency for 3 days were used in the influx assay; the roots were supplied with 5–200 lM 15Nurea for 6 min. 15N-urea uptake by the roots of ZmDUR3-transformed atdur3-3 mutant was saturable at *100 lM external urea, and showed a typical Michaelis–Menten kinetics with a Km of 9.429 ± 0.816 lM for urea and a Vmax of 1.286 ± 0.181 lmol g-1 DW h-1. Values are differences of 15N-urea of atdur3-3 mutant transformed with or without ZmDUR3 (atdur3-3 ? ZmDUR3-2 line was used, Fig. 8). Mean values ± SE (n = 5–7 biological replicates) were shown. 15N incorporated in the roots was analyzed by the mass spectrometry and converted into urea taken up by the root (see Materials and methods). DW root dry weight

Identification of a full-length cDNA encoding urea permease ZmDUR3 increases our understanding of maize functional genomics. Despite the availability of the sequenced maize genome (http://www.maizegdb.org; Schnable et al. 2009) and some valuable in silico analysis methods for maize DUR3 annotation (e.g. AC202439.3_FGP006 from ‘‘MaizeSequence’’ or ‘‘Aramemnon’’; AFW76805.1 from ‘‘NCBI’’; K7VFR8 from ‘‘UniProt’’), there seems to be no correct cDNA versions predicted for a functional maize DUR3 compared with the sequence isolated in this study (see Discussion below). Using a gene-prediction web service (‘‘CBS’’), RT-PCR and sequencing, we successfully cloned a cDNA with 2,196 bp for ZmDUR3 (Fig. 1), which was spliced from 4 exons of the maize genome and predicted to encode an integral membrane protein with 731 amino acid residues, 15 transmembrane domains and had 82.1 % sequence identity to OsDUR3 (Fig. 1). Inspection of maize EST databases revealed seven cDNA clones (acc. FL011289, FL011290, FL448872, AW400387, BQ163822, BQ163839, DV550376), all of whose sequences completely match to either the 50 - or 30 -end sequence across ‘‘start codon’’ or ‘‘stop codon’’ of ZmDUR3 and some of which exhibit a poly-A tail, implying that our isolated 2,196 bp cDNA should represent a full-length coding sequence for maize urea transporter ZmDUR3. Indeed, the molecular function of ZmDUR3 in urea transport or uptake was demonstrated by two independent

approaches using different systems. Firstly, a yeast functional complementation assay showed that the expression of ZmDUR3 could restore the growth phenotype of the YNVW1 mutant on 2 mM urea as a sole N source (Fig. 2), indicating a role of ZmDUR3 in urea import across the yeast plasma membrane (PM). Although the growth rate of the ZmDUR3-expressing yeast mutant on urea seemed not to be much affected by external pH in a range of 5–7 tested (Fig. 2), the complementation did occur under an acidic environment (Fig. 2). Moreover, the yeast periplasmic space is normally 2–3 pH units below that of the medium, which may influence the transport properties of heterologously expressed ZmDUR3 (Weidinger et al. 2007). Therefore, we cannot rule out that the urea transport mediated by ZmDUR3 could be coupled to protons, which is similar to the situation of AtDUR3 and OsDUR3 when expressed in yeast or Xenopus oocytes (Liu et al. 2003a; Wang et al. 2012). Secondly, using a transgenic approach, ZmDUR3-transformed atdur3-3 Arabidopsis lines were shown to be able to take up 50 % more externally supplied 15 N-urea than atdur3-3 mutant after exposure of plant roots to urea at a concentration of 50 or 500 lM for 20 min (Fig. 8c). Importantly, ZmDUR3-mediated urea influx into Arabidopsis roots occurred with Michealis–Menten kinetics with a Km * 9 lM (Fig. 7), consistent with a highaffinity transport process. In this work, we used the plant directly to successfully characterize the kinetic properties of a urea transporter ZmDUR3. Such data should, therefore, more closely reflect the molecular physiological process of urea movement actually undertaken by ZmDUR3 in its native plant. The function of ZmDUR3 is closely associated with nitrogen-nutritional status of maize The physiological relevance of ZmDUR3 in maize N nutrition, as well as in response to the external occurrence of varied N forms, including its putative substrate urea, was examined at a transcriptional level. The abundance of ZmDUR3 mRNA in roots was greatly and progressively increased over the time upon N-deprivation (Fig. 4), implying that ZmDUR3 should be regarded at least as a N-nutritional status responsive gene,particularly in the case of N-limitation. This scenario reflects a common feature of many N-form transporters including members of other DUR3 orthologs, AMT1 and NRT2 previously reported in certain unicellular organisms and higher plants when subjected to N deficiency (Hole et al. 1990; Glass et al. 2012; Sonoda et al. 2003; Kojima et al. 2007; Morel et al. 2008; Abreu et al. 2010; Yan et al. 2011; Wang et al. 2012; Andrews et al. 2013). However, the upregulation of ZmDUR3 induced by N deficiency was markedly decreased to near the control levels when resupplied for 3 h with urea

123

Planta

(Fig. 4). This contrasts with AtDUR3 and OsDUR3, where their expression is substrate (i.e., urea) inducible, even when N-starved and mRNA levels are already high (Kojima et al. 2007; Wang et al. 2012). This expression pattern of ZmDUR3 responsive to urea is reminiscent of other DUR3 homologs, e.g., PiDUR3 (from Aspergillus nidulans) and AnUreA (from Paxillus involutus), whose transcription was also not induced by their substrate urea or its precursors but subjected to N metabolite repression (Morel et al. 2008; Abreu et al. 2010). A similar phenomenon was described also for some ammonium- and nitrate-transporter genes in Arabidopsis and rice, e.g., the transcript abundance of OsAMT1;3, AtNRT1.3 and AtNRT2.5 in the roots occurred at highest levels under plant N-deprivation and thereafter was downregulated upon the resupply of their corresponding substrate NH4? or NO3- (Sonoda et al. 2003; Okamoto et al. 2003). Moreover, the transcriptional repression of ZmDUR3 by other N forms (NH4?, NO3- and Gln Fig. 4) might result from a feedback regulation of N metabolites such as glutamine (Gln), a first product of ammonium assimilation in plant cells, suggesting that ZmDUR3 responds to the N status of the tissue rather than to the presence of individual N-compounds. Indeed, a very similar result was reported for OsAMT1;3. The expression of this gene was suppressed by endogenously elevated Gln and externally applied Gln or Asn as well as NO3- (Sonoda et al. 2003); additionally, nitrate treatment increased the endogenous levels of NH4?, Gln, Glu, Asp and Asn (Sivasankar et al. 1997), which could further support the suggestion of the downregulation of ZmDUR3 by nitrate via downstream N metabolism. Thus, N-dependent expression studies provided strong evidence that ZmDUR3 as a urea transporter represents an important genetic component closely linking to plant N nutritional status. ZmDUR3 may represent a critical genetic determinant required for the urea-N remobilisation within maize plants In the case of nutrient deficiency or a high nutrient requirement, particularly during the reproductive stage, the uptake of N from the soil may not be sufficient for the plant. In this scenario, N-remobilization and reuse in the form of urea generated by the breakdown of N-containing macromolecules such as proteins and nucleotides from senescent tissues is thought to be prominent (Wang et al. 2008; Witte 2011). A positive correlation between higher transcripts accumulation of ZmDUR3 and elevating urea content detected in old leaves (Fig. 6a, b) would indicate a physiological role for ZmDUR3 with a great potential for N-recycling from aged tissues to growing sites, e.g., seed/ grain formation. During senescence, unlike animals that excrete their urine with urea at millimolar concentration ranges as a waste product (Sands 2004), higher plants are able to recycle and reuse their tissue nitrogen (Marschner 1995). Since the concentration of urea in the old leaves was

123

about 50 % higher than in the young ones and measured at lower micromole ranges (i.e., [200 lM if converted from the urea content assayed) (Fig. 6a), export of urea as a N source out of a source tissue to developing part should be achieved by the action of ZmDUR3 that has a low Km for its substrate. This seems to hold true, because a strong knockdown expression of rice active urea transporter OsDUR3 by T-DNA led to a higher accumulation of urea in the old leaves of the mutant compared to the wild-type plant (Wang et al. 2012). In such a case, ZmDUR3 might be crucial for phloem urea-loading and/or -unloading facilitating urea-N-recycling and reuse within the plant. Indeed, besides the major targeting of the ZmDUR3 protein at the PM shown by GFP-fusion analysis in the transgenic Arabidopsis lines (Fig. 3), tissue-level detection of gene expression using in situ PCR revealed that ZmDUR3 transcripts did occur in vascular-tissue cells of maize leaves, mostly within the phloem cell region (Fig. 5). ZmDUR3 would provide a promising molecular target enabling plants to effectively capture low external urea from the soil To access the likely physiological importance of the maize high-affinity urea transporter, the growth phenotype of Arabidopsis plants transformed with ZmDUR3 grown with urea or other N sources on agar nutrient medium was tested. Similar to OsDUR3 (Wang et al. 2012), ZmDUR3expressing atdur3-3 mutant lines or its corresponding WT plants showed improved growth rate on low urea, e.g., at 0.5 mM (Fig. 7a, b), while the mutant atdur3-3 and WT plants displayed obviously stunted growth phenotype with brown leaves, which is a typical symptom of N deficiency due to the accumulation of anthocyanin (Marschner 1995; Fig. 7a, b). Since the action of urease activity (i.e., hydrolyzing urea into HN4? and CO2), which exists pervasively in many plant species including maize (Wang et al. 2008; Gheibi et al. 2009), makes urea-N readily accessible to the plants (Witte 2011), increased accumulation of urea absorbed by roots via ZmDUR3-overexpression should allow plants to grow better at submillimolar urea concentrations (Fig. 7b). Further measurements of plant 15N derived from external supply of 15 N-urea absorbed by the roots revealed that ZmDUR3 overexpression enhanced urea uptake in transgenic plants by about 50 % compared to control mutant lines (Fig. 7c). All plants with or without ZmDUR3 transformation grew similarly well under the provision of NH4NO3 as a N source, suggesting that ZmDUR3 expression did not influence plants use of external NH4NO3 (Fig. 7a, b). Under the supply of higher urea (e.g., 3 mM), no clear growth difference was observed for all plants, which suggests that besides ZmDUR3, urea uptake might be mainly achieved by some low-affinity transport system(s), e.g., perhaps by certain members of the aquaporin family as reported for

Planta

Arabidopsis and maize (Liu et al. 2003b; Kojima et al. 2007; Gu et al. 2012). Nevertheless, our data provide evidence that ZmDUR3 expression does improve plants effective use of urea as a N source at lower millimolar concentrations that are reported in some natural environments, e.g., soils (Gaudin et al. 1987), probably by an improved ability to capture urea. Our future research will focus on increasing (maize) crop N-use efficiency by engineering the urea transport process, e.g., using ZmDUR3 as a promising candidate. Taken together, our results gained from molecular isolation and comprehensively functional characterization of the ZmDUR3 gene provide convincing evidence for the existence of a genetic basis coding for a high-affinity urea transporter in maize plants. The transport of urea by ZmDUR3 in both yeast and Arabidopsis plants, its transcriptional regulation by nitrogen, its targeting at the PM, and the occurrence of ZmDUR3 mRNA associated with phloem tissues strongly suggested that ZmDUR3 represents an active urea transporter. Our data are consistent with ZmDUR3 having a role in the capture of urea as a N source, especially under N-deficiency conditions, and in N-remobilization/-reuse in the form of urea within the plant from senescent organs. Together with the observation that ZmDUR3 improves plant growth on low urea by increasing net urea uptake when expressed in Arabidopsis, ZmDUR3 may offer a potential and valuable target for biotechnological approaches toward improving N-fertilizer-use efficiency in agricultural crop production. Author contribution Dr. GW Liu contributed to the major experimental analyses and partial manuscript writing, AL Sun conducted gene cloning and expression in yeast, Dr. M Gilliham and Dr. A Athman contributed to in-situ-PCR gene expression assay, Dr. DQ Li was involved in experiment design and gene expression study, Dr. LH Liu contributed to the general experiment design and manuscript preparation. Acknowledgments This work was financially supported by the National Natural Science Foundation of China (NSFC) (No. 31070223, Y333ZA1D11), and the Doctoral Research Fund Project of the Ministry of Education (No. 20134320110015) as well as the Synergistic Innovation Centre for The Southern Grain and Oil Crops of China (2011-Program awarded to Hunan Agricultural University).

References Abreu C, Sanguinetti M, Amillis S, Ramon A (2010) UreA, the major urea/H? symporter in Aspergillus nidulans. Fungal Genet Biol 47:1023–1033 Andrews M, Lea PJ (2013) Our nitrogen ‘footprint’: the need for increased crop nitrogen use efficiency. Ann Appl Biol 163:165–169

Andrews M, Raven JA, Lea PJ (2013) Do plants need nitrate? The mechanisms by which nitrogen form affects plants. Ann Appl Biol 163:174–199 Cameron KC, Di HJ, Moir JL (2013) Nitrogen losses from the soil/ plant system: a review. Ann Appl Biol 162:145–173 Gaudin R, Dupuy J, Bournat P (1987) Suivi du contenue en azote de la solution du sol d’une rizie`re apre`s placement d’ure´e. Agron Trop 42:13–19 Glass ADM, Britto DT, Kaiser BN, Kinghorn JR, Kronzucker HJ, Kumar A, Okamoto M, Rawat S, Siddiqi MY, Unkles SE, Vidmar JJ (2012) The regulation of nitrate and ammonium transport systems in plants. J Exp Bot 53:864–885 Gheibi MN, Kholdebarin B, Ghanati F, Teimouri S, Niroomand N, Samavati M (2009) Urease activity in maize (ZEA MAIZE L. CV. 704) as affected by nickel and nitrogen sources. Iran J Sci Technol Trans A 33(A4):299–307 Gu RL, Chen XL, Zhou YL, Yuan LY (2012) Isolation and characterization of three maize aquaporin genes, ZmNIP2;1, ZmNIP2;4 and ZmTIP4;4 involved in urea transport. BMB Rep 45:96–101 Hole DJ, Emran AM, Fares Y, Drew MC (1990) Induction of nitrate transport in maize roots, and kinetics of influx, measured with nitrogen-13. Plant Physiol 93:642–647 Jones DL, Healey JR, Willett VB, Farrar JF, Hodge A (2005) Dissolved organic nitrogen uptake by plants—an important N uptake pathway. Soil Biol Biochem 37:413–423 Kojima S, Bohner A, Gassert B, Yuan L, von Wire´n N (2007) AtDUR3 represents the major transporter for high-affinity urea transport across the plasma membrane of nitrogen-deficient Arabidopsis roots. Plant J 52:30–40 Krogmeier MJ, Maccarty W, Bremner JM (1989) Phytotoxicity of foliar-applied urea. Proc Natl Acad Sci USA 86:8189–8191 Liu LH, Ludewig U, Frommer WB, von Wire´n N (2003a) AtDUR3 encodes a new type of high affinity urea H? symporter in Arabidopsis. Plant Cell 15:790–800 Liu LH, Ludewig U, Gassert B, Frommer WB, von Wire´n N (2003b) Urea transport by nitrogen-regulated tonoplast intrinsic proteins in Arabidopsis. Plant Physiol 133:1220–1228 Marschner H (1995) Mineral nutrition of higher plants, 2nd edn. Academic Press, London Merigout P, Lelandais M, Bitton F, Renou JP, Briand X, Meyer C, Daniel-Vedele F (2008) Physiological and transcriptomic aspects of urea uptake and assimilation in Arabidopsis plants. Plant Physiol 147:1225–1238 Morel M, Jacob C, Fitz M, Wipf D, Chalot M, Brun A (2008) Characterization and regulation of PiDur3, a permease involved in the acquisition of urea by the ectomycorrhizal fungus Paxillus involutus. Fungal Genet Biol 45:912–921 Munns R, James RA, Xu B, Athman A, Conn SJ, Jordans C, Byrt CS, Hare RA, Tyerman SD, Tester M, Plett D, Gilliham M (2012) Wheat grain yield on saline soils is improved by an ancestral Na? transporter gene. Nat Biotechnol 30:360–364 Rentsch D, Schmidt S, Tegeder M (2007) Transporters for uptake and allocation of organic nitrogen compounds in plants. FEBS Lett 581:2281–2289 Sands JM (2004) Renal urea transporters. Curr Opin Nephrol Hypertens 13:525–532 Schnable PS, Ware D, Fulton RS et al (2009) The B73 maize genome: complexity, diversity, and dynamics. Science 326:1112–1115 Sivasankar S, Rothstein S, Oaks A (1997) Regulation of the accumulation and reduction of nitrate by nitrogen and carbon metabolites in maize seedlings. Plant Physiol 114:583–589 Sonoda Y, Ikeda A, Saiki S, von Wire´n N, Yamaya T, Yamaguchi J (2003) Distinct expression and function of three ammonium transporter genes (OsAMT1;1-1;3) in rice. Plant Cell Physiol 44:726–734

123

Planta Tanaka M, Wallace Junp IS, Takano E, Roberts DM, Fujiwara T (2008) NIP6;1 is a boric acid channel for preferential transport of boron to growing shoot tissues in Arabidopsis. Plant Cell 20:2860–2875 Tilman D, Cassman KG, Matson PA, Naylor R, Polasky S (2002) Agricultural sustainability and intensive production practices. Nature 418:671–677 Wang WH, Ko¨hler B, Cao FQ, Liu LH (2008) Molecular and physiological aspects of urea transport in higher plants. Plant Sci 175:467–477 Wang WH, Ko¨hler B, Cao FQ, Liu GW, Gong YY, Sheng S, Song QC, Cheng XY, Garnett T, Okamoto M, Qin R, Mueller-Roeber B, Tester M, Liu LH (2012) Rice DUR3 mediates high-affinity urea transport and plays an effective role in improvement of urea acquisition and utilization when expressed in Arabidopsis. New Phytol 193:432–444 Wang WH, Liu GW, Cao FQ, Cheng XY, Liu BW, Liu LH (2013) Inadequate root uptake may represent a major component

123

limiting rice to use urea as sole nitrogen source for growth. Plant Soil 363:191–200 Weidinger K, Neuhauser B, Gilch S, Ludewig U, Meyer O, Schmidt I (2007) Functional and physiological evidence for a Rhesus-type ammonia transporter in Nitrosomonas europaea. FEMS Microbiol Lett 273:260–267 Wieczorke R, Krampe S, Weierstall T, Freidel K, Hollenberg CP, Boles E (1999) Concurrent knock-out of at least 20 transporter genes is required to block uptake of hexoses in Saccharomyces cerevisiae. FEBS Lett 464:123–128 Witte CP (2011) Urea metabolism in plants. Plant Sci 180:431–438 Yan M, Fan XR, Feng HM, Miller AJ, Shen QR, Xu GH (2011) Rice OsNAR2.1 interacts with OsNRT2.1, OsNRT2.2 and OsNRT2.3a nitrate transporters to provide uptake over high and low concentration ranges. Plant, Cell Environ 34:1360–1372 Yu B, Yang Q, Kraus TEC, Dahlgren RA, Anastasio C, Zasoski RJ (2002) Contribution of compounds to dissolved organic nitrogen in forest soils. Biogeochemistry 61:173–198